06e887312b
* Add basic doc build system with sphinx and tox * Update README.md for how-to build docs * Update .gitignore to exclude docs builds * Add index and move README for Noop tests framework * TODO split and add more details for how-to noop tests setup/use/create Inspired by https://github.com/openstack/openstack-manuals Partial-bug: #1483288 Change-Id: I982e0baaff5de95c7d0c4b3f098754a80676f0d0 Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com> |
||
---|---|---|
.. | ||
conf.py | ||
index.rst | ||
README.rst |
Fuel noop rspec tests
Using the fuel_noop_tests.sh wrapper util
In order to test just execute these commands:
export WORKSPACE=/tmp/fuel_noop_tests
mkdir -p $WORKSPACE
./utils/jenkins/fuel_noop_tests.sh
In order to run specific test and/or specific astute.yaml, you can set appropriate env variables. For example:
export NOOP_TEST='keystone/*'
export NOOP_YAMLS='/path/to/your/astute.yaml'
./utils/jenkins/fuel_noop_tests.sh
If you also want to store puppet logs in case of catalog compilation errors, please set PUPPET_LOGS_DIR env variable:
export PUPPET_LOGS_DIR=/tmp/puppet_error_logs
If you want to store all the delcarated File and Package resources, please set NOOP_SAVE_RESOURCES_DIR env variable:
export NOOP_SAVE_RESOURCES_DIR=/tmp/puppet_resources
Using the fuel_noop_tests.rb util directly
The tool provides an advanced functionality. Use the -h key to get the help.
Deployment data layer checks
Below are typical use cases for the fuel_noop_tests.rb tool to perform a data layer checks of a change-set (a patch) against the committed state of the data layer.
Ruby and puppet version to be used (optional)
rvm use ruby-1.9.3-p545
PUPPET_GEM_VERSION=3.4.0
PUPPET_VERSION=3.4.0
Initial data templates generation (preparing the committed state)
Generate all data templates of all specs of all deployment scenarios making a reset1 & update of librarian puppet before
./utils/jenkins/fuel_noop_tests.rb -Q -b -r -u
the same but only for a particular ap-proxi spec (use -S to get the full list)
./utils/jenkins/fuel_noop_tests.rb -Q -b -s api-proxy/api-proxy_spec.rb
the same but only for the particular ap-proxi spec and the particular deployment scenario (use -Y to get the full list)
./utils/jenkins/fuel_noop_tests.rb -Q -b -s api-proxy/api-proxy_spec.rb -y novanet-compute.yaml
Running the data regression checks against a change-set under test
Run checks against the committed state of the data templates and save failed cases as a replay file
./utils/jenkins/fuel_noop_tests.rb -q -b -A replay.log
Re-run the data regression checks against the committed state of the data templates using the replay file and skipping all of the globals templates being re-generated again (a handy shourtcut for the test time)
./utils/jenkins/fuel_noop_tests.rb -q -b -g -a replay.log
Confirming the data changes made to became a new committed state
Use the same patterns as if making the Initial data templates generation. Amend generated files to the commit (the change-set under test).
Use ./deployment/remove_modules.sh to forcibly remove external modules in order to re-fetch them by the -r parameter.↩︎